Data Scenario and Model Hypothesis

Standard fit report for fits of hierSCAL to DERPA data.

Data Scenario: 1954_allFleets

Model Hypothesis: baseModel

Species: Dover, English, Rock

Stocks: HSHG, QCS, WCVI

Final phase convergence diagnostics

Max Gradient: 0.0803445

Objective Function value: -668.9885703

Time to fit model: 96.8840167

Full complex

Spawning Biomass

Spawning biomass (red line), catch (grey bars), and scaled biomass indices (coloured points), for all stocks and species. Indices are scaled by the fleet catchability, then by the ratio of spawning biomass to vulnerable biomass.

Figure 1: Spawning biomass (red line), catch (grey bars), and scaled biomass indices (coloured points), for all stocks and species. Indices are scaled by the fleet catchability, then by the ratio of spawning biomass to vulnerable biomass.

Fits to indices

Standardised vulnerable biomass (coloured lines) and the scaled and standardised biomass indices they are fitting to (coloured points), for all stocks and species.

Figure 2: Standardised vulnerable biomass (coloured lines) and the scaled and standardised biomass indices they are fitting to (coloured points), for all stocks and species.

Standardised residuals for model fits to biomass indices (coloured points) and a loess smoother with a 20% confidence interval (coloured lines and grey regions), for all fleets, stocks, and species.

Figure 3: Standardised residuals for model fits to biomass indices (coloured points) and a loess smoother with a 20% confidence interval (coloured lines and grey regions), for all fleets, stocks, and species.

Recruitment

Age-1 recruitments for all species and stocks. Equilibrium unfished recruitment $R_0$ is indicated by the horizontal dashed line.

Figure 4: Age-1 recruitments for all species and stocks. Equilibrium unfished recruitment \(R_0\) is indicated by the horizontal dashed line.

Deviations from expected recruitment for all species and stocks.

Figure 5: Deviations from expected recruitment for all species and stocks.

Stock-recruit curves (solid lines) and modeled recruitments (grey points), gridded over species (columns) and stocks (rows).

Figure 6: Stock-recruit curves (solid lines) and modeled recruitments (grey points), gridded over species (columns) and stocks (rows).

Time-varying catchability

Time series of observation model catchability (log-scale, coloured points) for the commercial fleets and Hecate Strait Multispecies Assemblage survey. Lines show the smoothed trend using a Loess smoother

Figure 7: Time series of observation model catchability (log-scale, coloured points) for the commercial fleets and Hecate Strait Multispecies Assemblage survey. Lines show the smoothed trend using a Loess smoother

Fishing mortality

Estimates of fishing mortality from each fleet (coloured points and lines), gridded over species (columns) and stocks (rows). Fishing mortality rates are found using an iterative Newton-Rhapson solver conditioned on the observed catch.

Figure 8: Estimates of fishing mortality from each fleet (coloured points and lines), gridded over species (columns) and stocks (rows). Fishing mortality rates are found using an iterative Newton-Rhapson solver conditioned on the observed catch.

Modeled removals (blue points) using estimated fishing mortality compared to observed catches (open circles), gridded over species (columns) and stocks (rows).

Figure 9: Modeled removals (blue points) using estimated fishing mortality compared to observed catches (open circles), gridded over species (columns) and stocks (rows).

Growth (Probability of Length-at-age)

Probability curves of length-at-age for males (blue) and females (red). Curves show the probability of each length within an age group, and opacity of the lines increases with age.

Figure 10: Probability curves of length-at-age for males (blue) and females (red). Curves show the probability of each length within an age group, and opacity of the lines increases with age.

Selectivity

Selectivity-at-length for each fleet, gridded over species (columns) and stocks (rows).

Figure 11: Selectivity-at-length for each fleet, gridded over species (columns) and stocks (rows).

Selectivity-at-age for each fleet, for females only, gridded over species (columns) and stocks (rows).

Figure 12: Selectivity-at-age for each fleet, for females only, gridded over species (columns) and stocks (rows).

Reference Points

Yield Curves

Equilibrium yield curves as a function of fishing mortality rates, assuming all fishing mortality comes from the modern trawl fleet.

Figure 13: Equilibrium yield curves as a function of fishing mortality rates, assuming all fishing mortality comes from the modern trawl fleet.

Goldilocks Plots

Re-read Hilborn 2018 for the goldilocks plot

Table

Table 1: Biological reference points for each stock and species.
Species Stock B0 R0 M_m M_f h Bmsy Fmsy Umsy MSY SSB_T D_T Dmsy_T
Dover HSHG 17.48 12.66 0.18 0.12 0.78 6 0.22 0.2 0.88 11.4 0.65 1.9
Dover QCS 1.87 1.77 0.2 0.14 0.78 0.6 0.23 0.2 0.11 1.36 0.73 2.27
Dover WCVI 14.78 17.79 0.22 0.15 0.78 4.82 0.26 0.22 0.97 8.38 0.57 1.74
English HSHG 12.05 13.8 0.19 0.13 0.77 4.1 0.36 0.31 0.89 6.53 0.54 1.59
English QCS 0.51 0.65 0.21 0.15 0.78 0.17 0.34 0.29 0.04 0.4 0.78 2.35
English WCVI 0.88 1 0.19 0.14 0.78 0.29 0.34 0.29 0.07 0.5 0.57 1.72
Rock HSHG 18.92 35.58 0.36 0.26 0.76 6.44 0.37 0.3 1.41 16.89 0.89 2.62
Rock QCS 5.62 5.18 0.27 0.19 0.76 1.93 0.27 0.23 0.36 1.97 0.35 1.02
Rock WCVI 2.05 1.84 0.26 0.19 0.75 0.74 0.34 0.29 0.13 0.7 0.34 0.95

Recruitment correlations

Standardised recruitment deviations. Blue indicates a positive deviation, red indicates a negative deviation, and the area of each circle is proportional to the size of the deviation.

Figure 14: Standardised recruitment deviations. Blue indicates a positive deviation, red indicates a negative deviation, and the area of each circle is proportional to the size of the deviation.

Correlation matrix of recruitment deviations. Blue indicates a positive correlation, and red indicates a negative correlation.

Figure 15: Correlation matrix of recruitment deviations. Blue indicates a positive correlation, and red indicates a negative correlation.

Individual species

UNDER CONSTRUCTION

Compositional data

Growth

Optimisation performance

Phase fit table

Table 2: Optimisation performance of hierSCAL for each phase.
phase objFun maxGrad nPar convCode convMsg time mcmcTime
1 3.826092e+07 2.2792095 577 0 relative convergence (4) 0.6526833 NA
2 3.813291e+06 22.7052110 586 0 relative convergence (4) 2.7892500 NA
3 3.813291e+06 0.2074114 586 1 false convergence (8) 0.8856333 NA
4 2.331320e+02 0.0380538 850 0 relative convergence (4) 16.7420167 NA
5 2.273129e+02 0.1438713 859 0 relative convergence (4) 4.5881167 NA
6 2.269242e+02 0.1363925 868 0 relative convergence (4) 5.0078833 NA
7 1.199150e+02 0.3650526 871 0 relative convergence (4) 11.9255833 NA
8 -4.137565e+00 0.0049006 877 0 relative convergence (4) 9.6495500 NA
9 -5.348749e+02 0.3995446 903 0 relative convergence (4) 28.1858167 NA
10 -6.415717e+02 0.1746256 926 0 relative convergence (4) 8.7375833 NA
11 -6.689886e+02 0.0803445 944 0 relative convergence (4) 7.7199000 NA
RE NA NA NA NA NA NA NA